Statistics for optimal point prediction in natural images.

Statistics for optimal point prediction in natural images.
复制标题

DOI:
10.1167/11.12.14
复制
发表时间:
2011-10-19
期刊:
影响因子:
1.8
通讯作者:
Perry JS
Perry JS
中科院分区:
医学4区
文献类型:
--
作者:
Geisler WS;Perry JS

文献摘要

参考文献

被引文献

相似文献

感觉系统利用自然信号的统计结构,因此,理解生物感觉系统和创建人工感觉系统的基本目标是表征自然信号的统计结构。在这里,我们使用一个简单的条件矩方法来测量与三个基本视觉任务相关的自然图像统计:(i)估计丢失或遮挡的图像点,(ii)从低分辨率图像(“超分辨率”)估计高分辨率图像,以及(iii)估计丢失的颜色通道。我们使用条件矩方法,因为它使最小的不变性假设,可以应用于任意大的训练数据集,并提供(给定足够的训练数据)贝叶斯最优估计。测量结果揭示了复杂但系统的统计特性,可以利用这些特性来大幅提高这三项任务的性能,而不是使用一些标准的图像处理方法。因此,这些统计数据很可能被人类视觉系统利用。
Sensory systems exploit the statistical regularities of natural signals, and thus, a fundamental goal for understanding biological sensory systems, and creating artificial sensory systems, is to characterize the statistical structure of natural signals. Here, we use a simple conditional moment method to measure natural image statistics relevant for three fundamental visual tasks: (i) estimation of missing or occluded image points, (ii) estimation of a high-resolution image from a low-resolution image (“super resolution”), and (iii) estimation of a missing color channel. We use the conditional moment approach because it makes minimal invariance assumptions, can be applied to arbitrarily large sets of training data, and provides (given sufficient training data) the Bayes optimal estimators. The measurements reveal complex but systematic statistical regularities that can be exploited to substantially improve performance in the three tasks over what is possible with some standard image processing methods. Thus, it is likely that these statistics are exploited by the human visual system.
DOI: 10.1109/tip.2004.838691
发表时间: 2005-03-01
影响因子: 10.6
作者:
Hirakawa, K;Parks, TW
通讯作者: Parks, TW
DOI: 10.1016/s0042-6989(97)00121-1
发表时间: 1997-12-01
期刊: VISION RESEARCH
影响因子: 1.8
作者:
Bell, AJ;Sejnowski, TJ
通讯作者: Sejnowski, TJ
DOI: 10.1167/9.13.17
发表时间: 2009-01-01
期刊: JOURNAL OF VISION
影响因子: 1.8
作者:
Geisler, Wilson S.;Najemnik, Jiri;Ing, Almon D.
通讯作者: Ing, Almon D.
DOI: 10.1109/38.988747
发表时间: 2002-03-01
影响因子: 1.8
作者:
Freeman, WT;Jones, TR;Pasztor, EC
通讯作者: Pasztor, EC
DOI: 10.1167/5.5.5
发表时间: 2005-01-01
期刊: JOURNAL OF VISION
影响因子: 1.8
作者:
Hofer, H;Singer, B;Williams, DR
通讯作者: Williams, DR